British Journal ofHaematology, 1976,34, 421.
Evidence for a Splenic Release Factor of Platelets in Chronic Blood Loss Plasma of Rabbits A. H. WEINTRAUB, I. KHANAND S. KARPATKIN Department of Medicine, N e w York University Medical School, N e w York, U.S.A. (Received 3 0 March 1976; acceptedfor publication
22
April 1976)
SUMMARY. Rabbits subjected to chronic blood loss with iron supplementation raised their platelet count and megathrombocyte number 1.7x and 2.1 x above baseline, respectively. In similarly treated splenectomized animals, no significant response was obtained. When chronic blood loss plasma was injected into recipient animals there was a rise in platelet count and megathrombocyte number of 1.8 x and 3.4 x respectively compared to animals injected with control plasma. In similarly treated splenectomized animals no such rise was obtained; neither was there a rise in platelet count or megathrombocyte number in intact recipient animals when chronic blood loss plasma was obtained from splenectomizedanimals. It is concluded that the thrombocytosis of chronic blood loss requires the presence of an intact spleen. The data suggest the presence of a release factor which requires the spleen for its elaboration as well as expression. The purpose of this investigation was to study the mechanism of thrombocytosis of blood loss (Richardson, 1904; Ode11 et al, 1962; Kelemen et a!, 1963 ; De Gabriele & Penington, 1967; Garg et al, 1972; Choi & Simone, 1973; Karpatkin et al, 1974; Karpatkin, 1974). Previous studies from our laboratory have revealed that chronic blood loss with iron supplementation leads to an elevation of platelet count and megathrombocyte number of 2.5 x and 3 . 8 ~in guinea-pigs (Garg et al, 1972; Karpatkin et al, 1974). Similar observations of 1.7x and 2.3 x respectively were made in rabbits (Karpatkin, 1974). The present investigation was stimulated by the observation that splenectomized animals subjected to chronic blood loss did not develop a significant thrombocytosis or megathrombocytosis. The present report presents evidence for the presence of a platelet release factor which requires the spleen for its elaboration as well as expression. MATERIALS AND METHODS New Zealand white rabbits of both sexes, weighing 3-4 kg, were employed. Splenectomies were performed under light intravenous sodium pentobarbital anaesthesia. The animals were given an intravenous injection of 500 ooo potassium penicillin G immediately after surgery, and nothing by mouth for the next 24 h. They were allowed to recover for 3-4 weeks before use. Correspondence: Dr S. Karpatkin; Department of Medicine, New York University Medical School, 550 First Avenue, New York, N.Y. 10016,U.S.A.
421
422
A. H. Weintraub, I. Khan and S . Karpatkin
Blood samples were collected into EDTA ‘vacutainer’ test tubes (Beckton-Dickinson & Co., Rutherford, N.J.) or ACD-A anticoagulant. Platelet counts were performed manually under phase-contrast microscopy (Brecher & Cronkite, 1950)~employing 3 % procaine hydrochloride as diluent (Karpatkin et a!, 1971). Megathrombocyte number was obtained by multiplying the proportion of megathrombocytes by the platelet count. The proportion of megathrombocytes was measured with a Coulter Counter, Model B attached to a P64 Channel Analyzer and 70 pm aperture tube (Coulter Electronics, Hialeah, Florida) from platelet-rich plasma, obtained by centrifugation in plastic 2 mm diameter tubes at 600 g for 30 s at room temperature, or by gravity sedimentation for 3 0 min at room temperature as described previously (Weintraub & Karpatkin, 1974). The platelet-rich plasma was diluted in cuvettes containing Isoton (Coulter Electronics) at room temperature. All counts were performed in duplicate and background counts were negligible. Each threshold division was made equal to 0.25 fl, so that windows 5-100 inclusive represented 1.25-25 fl volumes, which were taken to be 100% of the platelets. Windows 1-4 inclusive were eliminated because of interference by electrical ‘noise’. The median rabbit platelet volume ranged from 4 to fl and averaged 4.6 fl. Megathrombocytes were arbitrarily defined as those platelets in the upper 10% distribution volume. This corresponded to an average size range of 14-25 fl. The proportion of megathrombocytes was calculated by dividing the number of platelets in the 14-25 fl range by the total number of platelets (1.25-25 fl) and multiplying by 100. Experiments were in two groups: chronic blood loss in intact vs splenectomized animals; and injection of chronic blood loss plasma vs normal plasma into normal or splenectomized recipients. Chronic blood loss was initiated by the removal of approximately 15% blood volume every 3-4 d (assuming the intravascular volume to be 7% of body weight). The first blood withdrawn was discarded and the ensuing three or four collections combined. Iron loss was calculated and replaced with Imferon. The shed blood was collected into ACD-A anticoagulant (9 vols blood plus I vol anticoagulant). Plasma was obtained by centrifugation at 3000 rpm for 20 min in a Sorvall RC-3 Centrifuge and stored at - 20°C in 8 ml aliquots. Chronic blood loss plasma was administered intravenously into the lateral ear vein of recipient animals, 4-5 ml twice daily for 4 d. Control plasma obtained from a single blood withdrawal was similarly injected into other recipient animals. Statistical analysis was performed by the Student t-test. RESULTS Fig I demonstrates the effect of chronic blood loss with iron supplementation in intact vs splenectomized rabbits. The platelet count rose above basal levels in the intact rabbits to 1.3 x on day 3 ( P < O . O ~ )1.4x , on day 6 (P, 1.7x on day 10 (Po.I), 2.1 x on day 10 (P
Evidence for a Splenic Release Factor of Platelets in Chronic Blood Loss Plasma of Rabbits A. H. WEINTRAUB, I. KHANAND S. KARPATKIN Department of Medicine, N e w York University Medical School, N e w York, U.S.A. (Received 3 0 March 1976; acceptedfor publication
22
April 1976)
SUMMARY. Rabbits subjected to chronic blood loss with iron supplementation raised their platelet count and megathrombocyte number 1.7x and 2.1 x above baseline, respectively. In similarly treated splenectomized animals, no significant response was obtained. When chronic blood loss plasma was injected into recipient animals there was a rise in platelet count and megathrombocyte number of 1.8 x and 3.4 x respectively compared to animals injected with control plasma. In similarly treated splenectomized animals no such rise was obtained; neither was there a rise in platelet count or megathrombocyte number in intact recipient animals when chronic blood loss plasma was obtained from splenectomizedanimals. It is concluded that the thrombocytosis of chronic blood loss requires the presence of an intact spleen. The data suggest the presence of a release factor which requires the spleen for its elaboration as well as expression. The purpose of this investigation was to study the mechanism of thrombocytosis of blood loss (Richardson, 1904; Ode11 et al, 1962; Kelemen et a!, 1963 ; De Gabriele & Penington, 1967; Garg et al, 1972; Choi & Simone, 1973; Karpatkin et al, 1974; Karpatkin, 1974). Previous studies from our laboratory have revealed that chronic blood loss with iron supplementation leads to an elevation of platelet count and megathrombocyte number of 2.5 x and 3 . 8 ~in guinea-pigs (Garg et al, 1972; Karpatkin et al, 1974). Similar observations of 1.7x and 2.3 x respectively were made in rabbits (Karpatkin, 1974). The present investigation was stimulated by the observation that splenectomized animals subjected to chronic blood loss did not develop a significant thrombocytosis or megathrombocytosis. The present report presents evidence for the presence of a platelet release factor which requires the spleen for its elaboration as well as expression. MATERIALS AND METHODS New Zealand white rabbits of both sexes, weighing 3-4 kg, were employed. Splenectomies were performed under light intravenous sodium pentobarbital anaesthesia. The animals were given an intravenous injection of 500 ooo potassium penicillin G immediately after surgery, and nothing by mouth for the next 24 h. They were allowed to recover for 3-4 weeks before use. Correspondence: Dr S. Karpatkin; Department of Medicine, New York University Medical School, 550 First Avenue, New York, N.Y. 10016,U.S.A.
421
422
A. H. Weintraub, I. Khan and S . Karpatkin
Blood samples were collected into EDTA ‘vacutainer’ test tubes (Beckton-Dickinson & Co., Rutherford, N.J.) or ACD-A anticoagulant. Platelet counts were performed manually under phase-contrast microscopy (Brecher & Cronkite, 1950)~employing 3 % procaine hydrochloride as diluent (Karpatkin et a!, 1971). Megathrombocyte number was obtained by multiplying the proportion of megathrombocytes by the platelet count. The proportion of megathrombocytes was measured with a Coulter Counter, Model B attached to a P64 Channel Analyzer and 70 pm aperture tube (Coulter Electronics, Hialeah, Florida) from platelet-rich plasma, obtained by centrifugation in plastic 2 mm diameter tubes at 600 g for 30 s at room temperature, or by gravity sedimentation for 3 0 min at room temperature as described previously (Weintraub & Karpatkin, 1974). The platelet-rich plasma was diluted in cuvettes containing Isoton (Coulter Electronics) at room temperature. All counts were performed in duplicate and background counts were negligible. Each threshold division was made equal to 0.25 fl, so that windows 5-100 inclusive represented 1.25-25 fl volumes, which were taken to be 100% of the platelets. Windows 1-4 inclusive were eliminated because of interference by electrical ‘noise’. The median rabbit platelet volume ranged from 4 to fl and averaged 4.6 fl. Megathrombocytes were arbitrarily defined as those platelets in the upper 10% distribution volume. This corresponded to an average size range of 14-25 fl. The proportion of megathrombocytes was calculated by dividing the number of platelets in the 14-25 fl range by the total number of platelets (1.25-25 fl) and multiplying by 100. Experiments were in two groups: chronic blood loss in intact vs splenectomized animals; and injection of chronic blood loss plasma vs normal plasma into normal or splenectomized recipients. Chronic blood loss was initiated by the removal of approximately 15% blood volume every 3-4 d (assuming the intravascular volume to be 7% of body weight). The first blood withdrawn was discarded and the ensuing three or four collections combined. Iron loss was calculated and replaced with Imferon. The shed blood was collected into ACD-A anticoagulant (9 vols blood plus I vol anticoagulant). Plasma was obtained by centrifugation at 3000 rpm for 20 min in a Sorvall RC-3 Centrifuge and stored at - 20°C in 8 ml aliquots. Chronic blood loss plasma was administered intravenously into the lateral ear vein of recipient animals, 4-5 ml twice daily for 4 d. Control plasma obtained from a single blood withdrawal was similarly injected into other recipient animals. Statistical analysis was performed by the Student t-test. RESULTS Fig I demonstrates the effect of chronic blood loss with iron supplementation in intact vs splenectomized rabbits. The platelet count rose above basal levels in the intact rabbits to 1.3 x on day 3 ( P < O . O ~ )1.4x , on day 6 (P, 1.7x on day 10 (Po.I), 2.1 x on day 10 (P
